netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Daniel Borkmann <daniel@iogearbox.net>
To: ast@kernel.org
Cc: netdev@vger.kernel.org, Daniel Borkmann <daniel@iogearbox.net>
Subject: [PATCH bpf-next 5/6] bpf, x64: save 5 bytes in prologue when ebpf insns came from cbpf
Date: Sat, 24 Feb 2018 01:08:02 +0100	[thread overview]
Message-ID: <c28d794eceba4af88fb7517f892d255eca8b4857.1519430515.git.daniel@iogearbox.net> (raw)
In-Reply-To: <cover.1519430515.git.daniel@iogearbox.net>
In-Reply-To: <cover.1519430515.git.daniel@iogearbox.net>

While it's rather cumbersome to reduce prologue for cBPF->eBPF
migrations wrt spill/fill for r15 which is callee saved register
due to bpf_error path in bpf_jit.S that is both used by migrations
as well as native eBPF, we can still trivially save 5 bytes in
prologue for the former since tail calls can never be used there.
cBPF->eBPF migrations also have their own custom prologue in BPF
asm that xors A and X reg anyway, so it's fine we skip this here.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
 arch/x86/net/bpf_jit_comp.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 5b8fc13..70f9748 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -216,7 +216,7 @@ struct jit_context {
 /* emit x64 prologue code for BPF program and check it's size.
  * bpf_tail_call helper will skip it while jumping into another program
  */
-static void emit_prologue(u8 **pprog, u32 stack_depth)
+static void emit_prologue(u8 **pprog, u32 stack_depth, bool ebpf_from_cbpf)
 {
 	u8 *prog = *pprog;
 	int cnt = 0;
@@ -251,18 +251,21 @@ static void emit_prologue(u8 **pprog, u32 stack_depth)
 	/* mov qword ptr [rbp+24],r15 */
 	EMIT4(0x4C, 0x89, 0x7D, 24);
 
-	/* Clear the tail call counter (tail_call_cnt): for eBPF tail calls
-	 * we need to reset the counter to 0. It's done in two instructions,
-	 * resetting rax register to 0 (xor on eax gets 0 extended), and
-	 * moving it to the counter location.
-	 */
+	if (!ebpf_from_cbpf) {
+		/* Clear the tail call counter (tail_call_cnt): for eBPF tail
+		 * calls we need to reset the counter to 0. It's done in two
+		 * instructions, resetting rax register to 0, and moving it
+		 * to the counter location.
+		 */
 
-	/* xor eax, eax */
-	EMIT2(0x31, 0xc0);
-	/* mov qword ptr [rbp+32], rax */
-	EMIT4(0x48, 0x89, 0x45, 32);
+		/* xor eax, eax */
+		EMIT2(0x31, 0xc0);
+		/* mov qword ptr [rbp+32], rax */
+		EMIT4(0x48, 0x89, 0x45, 32);
+
+		BUILD_BUG_ON(cnt != PROLOGUE_SIZE);
+	}
 
-	BUILD_BUG_ON(cnt != PROLOGUE_SIZE);
 	*pprog = prog;
 }
 
@@ -453,7 +456,8 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
 	int proglen = 0;
 	u8 *prog = temp;
 
-	emit_prologue(&prog, bpf_prog->aux->stack_depth);
+	emit_prologue(&prog, bpf_prog->aux->stack_depth,
+		      bpf_prog_was_classic(bpf_prog));
 
 	if (seen_ld_abs)
 		emit_load_skb_data_hlen(&prog);
-- 
2.9.5

  parent reply	other threads:[~2018-02-24  0:08 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-24  0:07 [PATCH bpf-next 0/6] Few x64 jit improvements to shrink image size Daniel Borkmann
2018-02-24  0:07 ` [PATCH bpf-next 1/6] bpf, x64: save one byte per shl/shr/sar when imm is 1 Daniel Borkmann
2018-02-24  0:07 ` [PATCH bpf-next 2/6] bpf, x64: save several bytes by using mov over movabsq when possible Daniel Borkmann
2018-02-24  0:08 ` [PATCH bpf-next 3/6] bpf, x64: save several bytes when mul dest is r0/r3 anyway Daniel Borkmann
2018-02-24  0:08 ` [PATCH bpf-next 4/6] bpf, x64: save few bytes when mul is in alu32 Daniel Borkmann
2018-02-24  0:08 ` Daniel Borkmann [this message]
2018-02-24  0:08 ` [PATCH bpf-next 6/6] bpf: add various jit test cases Daniel Borkmann
2018-02-24  7:01 ` [PATCH bpf-next 0/6] Few x64 jit improvements to shrink image size Alexei Starovoitov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c28d794eceba4af88fb7517f892d255eca8b4857.1519430515.git.daniel@iogearbox.net \
    --to=daniel@iogearbox.net \
    --cc=ast@kernel.org \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).